Children acquiring English make systematic errors on subject control sentences even after they have reached near-adult competence (C. Chomsky, 1969), possibly due to heuristics based on semantic roles (Maratsos, 1974). Given the advanced fluency of large generative language models, we ask whether model outputs are consistent with these heuristics, and to what degree different models are consistent with each other. We find that models can be categorized by behavior into three separate groups, with broad differences between the groups. The outputs of models in the largest group are consistent with positional heuristics that succeed on subject control but fail on object control. This result is surprising, given that object control is orders of magnitude more frequent in the text data used to train such models. We examine to what degree the models are sensitive to prompting with agent-patient information, finding that raising the salience of agent and patient relations results in significant changes in the outputs of most models. Based on this observation, we leverage an existing dataset of semantic proto-role annotations (White, et al. 2020) to explore the connections between control and labeling event participants with properties typically associated with agents and patients.

最近已被证明大型语言模型在各种任务集中获得合理的零射普通化（Brown等，2020）。它已经假设这是语言模型的隐式多任务学习的结果，在语言模型中的预押（Radford等，2019）。可以通过明确的多任务学习直接引起零拍常规化？为了以缩放测试这个问题，我们开发一个系统，以便轻松地将任何自然语言任务映射到人类可读的提示表单中。我们转换一组大量的监督数据集，每个数据集都有多个提示，具有不同的措辞。这些提示的数据集允许基准测试模型执行完全看不见的任务的能力。我们介绍了一个普拉克尔编码器 - 解码器模型（Raffel等，2020; Lester等，2021），覆盖各种任务。该模型在多个标准数据集中达到强大的零点性能，通常优于其尺寸的型号超过16倍。此外，我们的方法对来自Big-替补基准测试的任务子集具有强烈性能，优于其尺寸的6倍。所有提示和培训的型号都可以在https://github.com/ bigscience-workshop / protectsource / httpsource / https：//huggingface.co/bigscience/t0pp。

Finetuning language models on a collection of datasets phrased as instructions has been shown to improve model performance and generalization to unseen tasks. In this paper we explore instruction finetuning with a particular focus on (1) scaling the number of tasks, (2) scaling the model size, and (3) finetuning on chain-of-thought data. We find that instruction finetuning with the above aspects dramatically improves performance on a variety of model classes (PaLM, T5, U-PaLM), prompting setups (zero-shot, few-shot, CoT), and evaluation benchmarks (MMLU, BBH, TyDiQA, MGSM, open-ended generation). For instance, Flan-PaLM 540B instruction-finetuned on 1.8K tasks outperforms PALM 540B by a large margin (+9.4% on average). Flan-PaLM 540B achieves state-of-the-art performance on several benchmarks, such as 75.2% on five-shot MMLU. We also publicly release Flan-T5 checkpoints, which achieve strong few-shot performance even compared to much larger models, such as PaLM 62B. Overall, instruction finetuning is a general method for improving the performance and usability of pretrained language models.

本文探讨了提高语言模型的零次学习能力的简单方法。我们表明，指令调整 - 通过对说明书中所述的任务集合微调语言模型 - 大幅提升零射门上看不见任务中的表现。我们采取预训练的语言模型和指令调整它通过自然语言指令模板语言表达了60NLP任务137B参数。我们评估这种指令调整模型，我们称之为FLAN，在看不见的任务类型。FLAN显着改善其未修饰的对应的性能和超过25的20个任务，我们评估零射门175BGPT-3。FLAN甚至GPT-3通过在安利，RTE，BoolQ，AI2-ARC，OpenbookQA和StoryCloze大比分胜过几拍。消融研究显示任务和模型的规模，这个数字是指令调整取得成功的关键组成部分。

Instruction tuning enables pretrained language models to perform new tasks from inference-time natural language descriptions. These approaches rely on vast amounts of human supervision in the form of crowdsourced datasets or user interactions. In this work, we introduce Unnatural Instructions: a large dataset of creative and diverse instructions, collected with virtually no human labor. We collect 64,000 examples by prompting a language model with three seed examples of instructions and eliciting a fourth. This set is then expanded by prompting the model to rephrase each instruction, creating a total of approximately 240,000 examples of instructions, inputs, and outputs. Experiments show that despite containing a fair amount of noise, training on Unnatural Instructions rivals the effectiveness of training on open-source manually-curated datasets, surpassing the performance of models such as T0++ and Tk-Instruct across various benchmarks. These results demonstrate the potential of model-generated data as a cost-effective alternative to crowdsourcing for dataset expansion and diversification.

预测任务标签和为其预测生成自由文本阐述的自律化模型可以实现与NLP系统更直观的交互。然而，这些模型目前正在接受大量人为的自由文本解释，每个任务都会阻碍更广泛的使用。我们建议使用少数培训例子研究更现实的自律化建立。我们出示2月 - 一个标准化的四个现有英语数据集和相关指标。我们通过2月份广泛探索自然语言提示来确定正确的提示方法。然后，通过使用此提示并缩放模型大小，我们证明了几次拍摄自合合理化的进展。我们展示了这项任务的完善房间仍然有充足的改进空间：人类注册人评估的生成解释的平均合理性最多为51％，而人类解释的合理性是76％。我们希望2月份与我们的拟议方法一起促使社区承担几次拍摄的自我合理化挑战。

培训和评估语言模型越来越多地要求构建元数据 - 多样化的策划数据收集，并具有清晰的出处。自然语言提示最近通过将现有的，有监督的数据集转换为多种新颖的预处理任务，突出了元数据策划的好处，从而改善了零击的概括。尽管将这些以数据为中心的方法转化为生物医学语言建模的通用域文本成功，但由于标记的生物医学数据集在流行的数据中心中的代表性大大不足，因此仍然具有挑战性。为了应对这一挑战，我们介绍了BigBio一个由126个以上的生物医学NLP数据集的社区库，目前涵盖12个任务类别和10多种语言。 BigBio通过对数据集及其元数据进行程序化访问来促进可再现的元数据策划，并与当前的平台兼容，以及时工程和端到端的几个/零射击语言模型评估。我们讨论了我们的任务架构协调，数据审核，贡献指南的过程，并概述了两个说明性用例：生物医学提示和大规模，多任务学习的零射门评估。 BigBio是一项持续的社区努力，可在https://github.com/bigscience-workshop/biomedical上获得。

GPT-3等大型自回归语言模型是几秒钟的学习者，可以在没有微调的情况下执行各种语言任务。虽然已知这些模型能够共同代表许多不同的语言，但他们的培训数据由英语主导，可能限制了它们的交叉概括。在这项工作中，我们在覆盖多种语言的平衡语料库上培训多语言自回归语言模型，并在广泛的任务中研究他们几乎没有零点的学习能力。我们最大的模型，具有75亿参数，在20多种代表语言中，在几种代表语言中，在几种代表性语言中，在几种代表性语言中，在多语言型号推理中表现出可比大小的GPT-3（在0次设置和0次拍摄设置中的绝对精度改善+ 7.4％ 4-拍摄设置中的9.4％）和自然语言推理（每次拍摄和4次设置中的每一个+ 5.4％）。在Flores-101机器翻译基准测试中，我们的模型优于GPT-3在182个翻译方向上有32个培训例子，同时超过45个方向的官方监督基线。我们介绍了模型成功和失败的位置的详细分析，特别是它尤其显示在某些任务中实现交叉语境的内容学习，而仍然存在改善表面的鲁棒性和适应没有a的任务的余地自然冻结形式。最后，我们评估我们在仇恨语音检测中以五种语言的仇恨语音检测的模型，并发现它具有与可比大小的GPT-3模型类似的限制。

Language models have recently achieved strong performance across a wide range of NLP benchmarks. However, unlike benchmarks, real world tasks are often poorly specified, and agents must deduce the user's intended behavior from a combination of context, instructions, and examples. We investigate how both humans and models behave in the face of such task ambiguity by proposing AmbiBench, a new benchmark of six ambiguously-specified classification tasks. We evaluate humans and models on AmbiBench by seeing how well they identify the intended task using 1) instructions with varying degrees of ambiguity, and 2) different numbers of labeled examples. We find that the combination of model scaling (to 175B parameters) and training with human feedback data enables models to approach or exceed the accuracy of human participants across tasks, but that either one alone is not sufficient. In addition, we show how to dramatically improve the accuracy of language models trained without large-scale human feedback training by finetuning on a small number of ambiguous in-context examples, providing a promising direction for teaching models to generalize well in the face of ambiguity.

大型语言模型在零拍设置中显示出令人鼓舞的结果（Brown等，2020； Radford等，2019）。例如，他们只需在问题上调节并以最高概率选择答案来执行多项选择任务。但是，由于表面竞争的表面形式 - 在不同的表面形式竞争概率质量，即使它们代表相同的基本概念，例如“计算机”和“ PC”。由于概率质量是有限的，因此由于其他是有效答案的字符串的竞争（但不是多项选择选项之一），这会降低正确答案的概率。我们引入域有条件地互相信息，这是一种替代评分函数，可以通过简单地根据特定的零击任务的上下文中的先验可能性重新重新拨出每个选项来直接补偿表面竞争。在校准（Zhao等，2021）和所有GPT-2和GPT-3模型上，在各种多项选择数据集上，它都可以在零击性能方面的一致增长和未校准的评分功能。

We demonstrate that scaling up language models greatly improves task-agnostic, few-shot performance, sometimes even becoming competitive with prior state-ofthe-art fine-tuning approaches. Specifically, we train GPT-3, an autoregressive language model with 175 billion parameters, 10x more than any previous nonsparse language model, and test its performance in the few-shot setting. For all tasks, GPT-3 is applied without any gradient updates or fine-tuning, with tasks and few-shot demonstrations specified purely via text interaction with the model. GPT-3 achieves strong performance on many NLP datasets, including translation, question-answering, and cloze tasks. We also identify some datasets where GPT-3's few-shot learning still struggles, as well as some datasets where GPT-3 faces methodological issues related to training on large web corpora.

基于变压器的语言模型最近在许多自然语言任务中取得了显着的结果。但是，通常通过利用大量培训数据来实现排行榜的性能，并且很少通过将明确的语言知识编码为神经模型。这使许多人质疑语言学对现代自然语言处理的相关性。在本文中，我介绍了几个案例研究，以说明理论语言学和神经语言模型仍然相互关联。首先，语言模型通过提供一个客观的工具来测量语义距离，这对语言学家很有用，语义距离很难使用传统方法。另一方面，语言理论通过提供框架和数据源来探究我们的语言模型，以了解语言理解的特定方面，从而有助于语言建模研究。本论文贡献了三项研究，探讨了语言模型中语法 - 听觉界面的不同方面。在论文的第一部分中，我将语言模型应用于单词类灵活性的问题。我将Mbert作为语义距离测量的来源，我提供了有利于将单词类灵活性分析为方向过程的证据。在论文的第二部分中，我提出了一种方法来测量语言模型中间层的惊奇方法。我的实验表明，包含形态句法异常的句子触发了语言模型早期的惊喜，而不是语义和常识异常。最后，在论文的第三部分中，我适应了一些心理语言学研究，以表明语言模型包含了论证结构结构的知识。总而言之，我的论文在自然语言处理，语言理论和心理语言学之间建立了新的联系，以为语言模型的解释提供新的观点。

As the performance of large language models rapidly improves, benchmarks are getting larger and more complex as well. We present LMentry, a benchmark that avoids this "arms race" by focusing on a compact set of tasks that are trivial to humans, e.g. writing a sentence containing a specific word, identifying which words in a list belong to a specific category, or choosing which of two words is longer. LMentry is specifically designed to provide quick and interpretable insights into the capabilities and robustness of large language models. Our experiments reveal a wide variety of failure cases that, while immediately obvious to humans, pose a considerable challenge for large language models, including OpenAI's latest 175B-parameter instruction-tuned model, TextDavinci002. LMentry complements contemporary evaluation approaches of large language models, providing a quick, automatic, and easy-to-run "unit test", without resorting to large benchmark suites of complex tasks.

大型语言模型越来越能够通过相对较少的特定任务的监督产生流畅的出现文本。但这些模型可以准确解释分类决策吗？我们考虑使用少量人写的例子（即，以几滴方式）生成自由文本解释的任务。我们发现（1）创作更高质量的例子，以提示导致更高质量的世代; （2）令人惊讶的是，在头到头比较中，人群公司通常更喜欢GPT-3生成的解释，以众包中包含的人性写入的解释。然而，Crowdworker评级也表明，虽然模型产生了事实，语法和充分的解释，但它们具有改进的空间，例如沿着提供新颖信息和支持标签的轴。我们创建了一种管道，该管道将GPT-3与监督过滤器结合起来，该过滤器通过二进制可接受性判断来包含人类循环。尽管具有重要的主观性内在的判断可接受性，但我们的方法能够始终如一地过滤人类可接受的GPT-3生成的解释。

Recent work has shown that fine-tuning large pre-trained language models on a collection of tasks described via instructions, a.k.a. instruction-tuning, improves their zero and few-shot generalization to unseen tasks. However, there is a limited understanding of the performance trade-offs of different decisions made during the instruction-tuning process. These decisions include the scale and diversity of the instruction-tuning benchmark, different task sampling strategies, fine-tuning with and without demonstrations, training using specialized datasets for reasoning and dialogue, and finally, the fine-tuning objectives themselves. In this paper, we characterize the effect of instruction-tuning decisions on downstream task performance when scaling both model and benchmark sizes. To this end, we create OPT-IML Bench: a large benchmark for Instruction Meta-Learning (IML) of 2000 NLP tasks consolidated into task categories from 8 existing benchmarks, and prepare an evaluation framework to measure three types of model generalizations: to tasks from fully held-out categories, to held-out tasks from seen categories, and to held-out instances from seen tasks. Through the lens of this framework, we first present insights about instruction-tuning decisions as applied to OPT-30B and further exploit these insights to train OPT-IML 30B and 175B, which are instruction-tuned versions of OPT. OPT-IML demonstrates all three generalization abilities at both scales on four different evaluation benchmarks with diverse tasks and input formats -- PromptSource, FLAN, Super-NaturalInstructions, and UnifiedSKG. Not only does it significantly outperform OPT on all benchmarks but is also highly competitive with existing models fine-tuned on each specific benchmark. We release OPT-IML at both scales, together with the OPT-IML Bench evaluation framework.

当前的语言模型可以产生高质量的文本。他们只是复制他们之前看到的文本，或者他们学习了普遍的语言抽象吗？要取笑这些可能性，我们介绍了乌鸦，这是一套评估生成文本的新颖性，专注于顺序结构（n-gram）和句法结构。我们将这些分析应用于四种神经语言模型（LSTM，变压器，变换器-XL和GPT-2）。对于本地结构 - 例如，单个依赖性 - 模型生成的文本比来自每个模型的测试集的人类生成文本的基线显着不那么新颖。对于大规模结构 - 例如，总句结构 - 模型生成的文本与人生成的基线一样新颖甚至更新颖，但模型仍然有时复制，在某些情况下，在训练集中重复超过1000字超过1,000字的通道。我们还表现了广泛的手动分析，表明GPT-2的新文本通常在形态学和语法中形成良好，但具有合理的语义问题（例如，是自相矛盾）。

我们建议一种使用大型语言模型（例如GPT-3）在给定情况下模拟不同人类的响应的方法。我们通过尝试重现公认的经济，心理语言和社会实验来测试我们的方法。该方法需要每个实验的及时模板。模拟是通过改变（假设的）主题细节（例如名称）和分析语言模型产生的文本来运行的。我们通过使用GPT-3来验证我们的方法，以表明可以模拟不同人的反应，并且他们的反应与文献中先前的人类研究一致。我们发现，较大的语言模型生成的分布与先前的实验结果更好地保持一致，这表明一种趋势，即未来的语言模型可以用于对人类反应的更忠实的模拟。我们将语言模型用于模拟的使用与对语言模型的拟人化观点形成对比。

Generating a chain of thought (CoT) can increase large language model (LLM) performance on a wide range of tasks. Zero-shot CoT evaluations, however, have been conducted primarily on logical tasks (e.g. arithmetic, commonsense QA). In this paper, we perform a controlled evaluation of zero-shot CoT across two sensitive domains: harmful questions and stereotype benchmarks. We find that using zero-shot CoT reasoning in a prompt can significantly increase a model's likelihood to produce undesirable output. Without future advances in alignment or explicit mitigation instructions, zero-shot CoT should be avoided on tasks where models can make inferences about marginalized groups or harmful topics.

People constantly use language to learn about the world. Computational linguists have capitalized on this fact to build large language models (LLMs) that acquire co-occurrence-based knowledge from language corpora. LLMs achieve impressive performance on many tasks, but the robustness of their world knowledge has been questioned. Here, we ask: do LLMs acquire generalized knowledge about real-world events? Using curated sets of minimal sentence pairs (n=1215), we tested whether LLMs are more likely to generate plausible event descriptions compared to their implausible counterparts. We found that LLMs systematically distinguish possible and impossible events (The teacher bought the laptop vs. The laptop bought the teacher) but fall short of human performance when distinguishing likely and unlikely events (The nanny tutored the boy vs. The boy tutored the nanny). In follow-up analyses, we show that (i) LLM scores are driven by both plausibility and surface-level sentence features, (ii) LLMs generalize well across syntactic sentence variants (active vs passive) but less well across semantic sentence variants (synonymous sentences), (iii) some, but not all LLM deviations from ground-truth labels align with crowdsourced human judgments, and (iv) explicit event plausibility information emerges in middle LLM layers and remains high thereafter. Overall, our analyses reveal a gap in LLMs' event knowledge, highlighting their limitations as generalized knowledge bases. We conclude by speculating that the differential performance on impossible vs. unlikely events is not a temporary setback but an inherent property of LLMs, reflecting a fundamental difference between linguistic knowledge and world knowledge in intelligent systems.

在这项工作中，我们证明了多种语的大规模序列到序列（SEQ2SEQ）模型，该模型是通过Denoising和因果语言建模（CLM）任务的混合物进行训练的，比仅解码器模型更有效地进行了效率的学习者在各种任务上。特别是，我们培训了一个名为Alexa教师模型（Alexatm 20b）的200亿个参数多语言SEQ2SEQ模型，并表明它在1-Shot摘要任务上实现了最先进的（SOTA）性能，超过了更大的540B PALM DOPODER模型。 Alexatm 20b还可以在1-Shot Machine翻译中实现SOTA，尤其是对于低资源语言，几乎所有语言对（阿拉伯语，英语，法语，德语，德语，印地语，意大利语，日语，以及flores-101数据集上的泰卢固语）。我们还显示了零拍设置，AlexATM 20B在SuperGlue和SqueadV2数据集上的表现优于GPT3（175B），并在XNLI，XCOPA，PAWS-X和XWINOGRAD等多语言任务上提供SOTA性能。总体而言，我们的结果为SEQ2SEQ模型提供了一个令人信服的案例，作为大型语言模型（LLM）培训的仅解码器模型的强大替代方法。